|
Cicero posted:Yeah, I'll take this bet. Define "can". Who cares about "can"? I think self-driving cars "can" do lots of things in 10 years except broadly impact the market or really shake up transportation in general.
|
#
?
Apr 9, 2016 02:17
|
|
|
|
| # ? May 22, 2024 18:18 |
|
|
Cicero posted:Yeah, I'll take this bet. I legitimately don't see this happening in 10 years. Taking the over on this one.  Now, on another transportation subject much closer to this thread's heart, there's a new women-only rideshare service coming out. Their press-release goes hard against Uber, too: Here's Why Women Everywhere Will Delete Uber On April 19. Every post I've seen sharing this article on Facebook has degenerated into weird hostile #notallmen / #yesallmen gender poo poo, but one interesting thing this brings up is that Uber basically has no "moat". Other than the matter of scale, there isn't much to keep other copycat services (with more specific targeting) from popping up to eat into their market share. Most drivers already drive for both Uber and Lyft, grabbing rides from whichever app happens to ping them first. I wouldn't be surprised if every driver for this service does the same thing and just runs it along with Uber and Lyft. In fact, Uber's scale could actually be helpful to these competitors, because Uber will have created a pool of drivers in each market who can add on driving for this service for essentially no extra cost to them. That will make driver recruitment, probably one of the hardest hurdles for such a service, much easier. I wonder if the VCs who managed to give Uber its $62.5 billion valuation considered that.
|
|
#
?
Apr 9, 2016 02:29
|
|
|
e_angst posted:I legitimately don't see this happening in 10 years. Taking the over on this one. Uber's biggest advantage at this point is network effects. These are what make Uber so convenient and fast. Guess what the downfall of an app that has only women drivers (about 15% of drivers, but probably well below that in terms of rides/miles driven) will be? Oh yeah, they also have all of Uber's regulatory problems, plus additional ones related to (explicit) discrimination, and none of the VC dollars necessary to fight them. blah_blah fucked around with this message at 02:46 on Apr 9, 2016 |
|
#
?
Apr 9, 2016 02:43
|
|
|
Yeah, that's wrong. The thing you have to understand about Uber is that it's a two sided platform. It not only has to attract enough drivers to service demand, it also has to attract enough riders to deliver enough fairs for it to be worth those driver's time. If either side is not in equilibrium, the platform fails. Because starting and growing a two sided platform is so hard, this is actually the kind of business where network effects are the strongest. There's a kind of virtuous cycle if you do it right where more drivers leads to better service which leads to more fares which attracts more drivers. Conversely, if a platform with that kind of network already exists in an industry, it becomes very hard for a competitor to gain a foothold, because the experience is going to be demonstrably worse for both sides in the beginning. The exception is if the new platform has some sort of tech or feature that allows it to provide such a better service that it's really not the same kind of platform anymore. But if it's just commodity tech against commodity tech, the bigger network always wins. Don't get me wrong, Uber has real problems. But getting competed or of existence by a bunch of scrub companies is not one.
|
|
#
?
Apr 9, 2016 03:19
|
|
|
theyre gonna DISRUPT DRIVING
|
|
#
?
Apr 9, 2016 03:32
|
|
|
Baby Babbeh posted:Yeah, that's wrong. The thing you have to understand about Uber is that it's a two sided platform. It not only has to attract enough drivers to service demand, it also has to attract enough riders to deliver enough fairs for it to be worth those driver's time. If either side is not in equilibrium, the platform fails. But what I'm saying is that some of the scrubs can basically borrow Uber's network effect because the only thing drivers have to do is download one more app to run while out driving (this is why, as I mentioned, pretty much every driver I know does both Uber and Lyft). The cost for the driver to be part of this is low so even if they only end up with one or two extra fares a night, it could be worth it. (Especially in areas where the number of drivers is very high.) How popular the service is with riders is definitely an important question, but I know I've seen the article I linked shared a lot among my female friends on Facebook.
|
|
#
?
Apr 9, 2016 03:50
|
|
|
Konstantin posted:I agree, simply because there is too much money to be made. Forget personal cars, the real money is in driverless semi trucks, taxis, shuttles, buses, delivery trucks, and other fleet vehicles. Getting humans out of the driver's seat would massively reduce transportation and shipping costs, and the big companies that stand to gain from it can push the relevant laws through. This is true, but at the same time, subway and train drivers. I mean, if we still haven't automated something that's basically a car trip that always go through the same path, with mostly the same obstacles I can't see us getting rid of truck drivers in 10 years.
|
|
#
?
Apr 9, 2016 04:35
|
|
|
Mr. Nemo posted:This is true, but at the same time, subway and train drivers. I mean, if we still haven't automated something that's basically a car trip that always go through the same path, with mostly the same obstacles I can't see us getting rid of truck drivers in 10 years. A lot of those drivers are there for liability reasons. Which is probably what will happen when we do get true driverless vehicles.
|
|
#
?
Apr 9, 2016 05:30
|
|
|
computer parts posted:A lot of those drivers are there for liability reasons. 
|
|
#
?
Apr 9, 2016 05:42
|
|
|
Truck convoys. 1 driver in the front truck and ten trucks without drivers that follow it. This Wednesday a Dutch government challenge to demonstrate a similar tech was succesfully completed. They had trucks from several brands follow each other very closely and brake/accelerate synchronized. https://www.theguardian.com/technology/2016/apr/07/convoy-self-driving-trucks-completes-first-european-cross-border-trip
|
|
#
?
Apr 9, 2016 07:36
|
|
|
Lucy Heartfilia posted:Truck convoys. 1 driver in the front truck and ten trucks without drivers that follow it. This Wednesday a Dutch government challenge to demonstrate a similar tech was succesfully completed. They had trucks from several brands follow each other very closely and brake/accelerate synchronized. So trains. They're trains.
|
|
#
?
Apr 9, 2016 08:06
|
|
|
Land-trains thank you kindly.
|
|
#
?
Apr 9, 2016 08:18
|
|
|
cheese posted:Which then becomes a question of "Why have a driverless big rig if it has to have a driver in it anyway?" This is kind of what I was going for. And the answer I've read is "safety measures". Human drivers are obviously not perfect, and neither are driveless vehicles. But you couls combine part of both systems. And have a driver based vehicle have some advanced collision detection software, where it allows a computer to react to certain situations, quicker than a human could. Or maybe the opposite, which means the human is there to take over in case of an emergency, kind of like trains and subways. But this just results in increased hardware costs which in turn increases delivery costs. Sure, you could end up with less dead people, but that doesn't improve the bottom line.
|
|
#
?
Apr 9, 2016 08:18
|
|
|
Mr. Nemo posted:This is true, but at the same time, subway and train drivers. I mean, if we still haven't automated something that's basically a car trip that always go through the same path, with mostly the same obstacles I can't see us getting rid of truck drivers in 10 years. One issue that metro systems in the US have is that the US just tends to underinvest in transit in general, combine that with union resistance and it's not surprising that our automated trains are just in little niches.
|
|
#
?
Apr 9, 2016 08:54
|
|
|
computer parts posted:A lot of those drivers are there for liability reasons. This is effectively the case for parts of the London Underground / Docklands Light Railway. Under normal operations drivers aren't pushing stop / go, however they are operating doors, monitoring passengers and are required to take over if anything goes wrong. (They're also pretty strongly unionised.) Part of the Underground's issue is that a substantial chunk of its tunnels are really closely intertwined with the cables, pipes and foundations of the city. This means that most lines don't have any kind of escapeway and there's zero possibility of expanding tunnels to add them. If a train failed between stations you'd absolutely need a person able to get there to guide passengers to safety, and right now having that person be the driver is the easiest solution to that.
|
|
#
?
Apr 9, 2016 09:28
|
|
|
Peztopiary posted:Land-trains thank you kindly. As opposed to...sky-trains?
|
|
#
?
Apr 9, 2016 13:26
|
|
|
The sf trains and busses are always failing in minor ways. The driver will just come and do some minor fix. Without people to do that the system would poo poo a brick.
|
|
#
?
Apr 9, 2016 14:05
|
|
|
Liquid Communism posted:Automated cars are a joke, and will be no closer to public adoption 10 years ago than they are now because they flatly aren't capable of fulfilling many of the jobs vehicles are used for, and are another high maintenance component with a strong liklihood of a malfunction killing drive/passenger/bystanders. You are 100% wrong on this. I have talked with product planners from the Big 3 US Automakers and they all plan on having a high-end self-driving car for 2020, and expect the technology to be throughout their product lines by 2025. Self-driving cars will be like the iPhone, they will break the logjam and the technology will be everywhere very quickly, leaving people wondering how they ever lived without it.
|
|
#
?
Apr 9, 2016 14:17
|
|
|
on the left posted:You are 100% wrong on this. I have talked with product planners from the Big 3 US Automakers and they all plan on having a high-end self-driving car for 2020, and expect the technology to be throughout their product lines by 2025. When you say "self-driving car" do you mean level 4 autonomy? I'm an autonomous vehicle booster for sure, but that seems pretty aggressive.
|
|
#
?
Apr 9, 2016 15:38
|
|
|
Subjunctive posted:When you say "self-driving car" do you mean level 4 autonomy? I'm an autonomous vehicle booster for sure, but that seems pretty aggressive. Tesla's autonomous driving functions are level 2 by the way. Several manufacturers will release level 3 autonomous driving cars in a few years.
|
|
#
?
Apr 9, 2016 15:52
|
|
|
Lucy Heartfilia posted:Truck convoys. 1 driver in the front truck and ten trucks without drivers that follow it. This Wednesday a Dutch government challenge to demonstrate a similar tech was succesfully completed. They had trucks from several brands follow each other very closely and brake/accelerate synchronized. Like what someone said earlier, maybe this will be a thing in Europe because they don't use trains for shipping cargo, but this is incredibly inefficient in the US.
|
|
#
?
Apr 9, 2016 16:51
|
|
|
Lucy Heartfilia posted:Tesla's autonomous driving functions are level 2 by the way. Several manufacturers will release level 3 autonomous driving cars in a few years. Yeah, I'm familiar with the Tesla ones. A few years from now is 2019. I think there's a big gap between 3 and 4 in terms of edge case handling, which will take more than a year to sort out.
|
|
#
?
Apr 9, 2016 17:00
|
|
|
Also that might be considered a convoy which is illegal in most states iirc
|
|
#
?
Apr 9, 2016 17:01
|
|
|
Mercy sakes alive.
|
|
#
?
Apr 9, 2016 17:12
|
|
|
moller posted:Mercy sakes alive.  It seems to me that long autonomous convoys are a safety hazard anyway, because of the crazy things the car drivers will do to get around the large blob in the right lane (and probably driving slowly, for braking's sake.)
|
|
#
?
Apr 9, 2016 17:17
|
|
|
If they have to drive slowly, the whole thing is useless. But I don't think they will, because the effective reaction time for the following trucks is zero, so they can stop as fast as the lead truck at any time.
|
|
#
?
Apr 9, 2016 17:22
|
|
|
A friend of a friend just launched his start-up. I feel a little guilty for posting it here because he seems like a nice enough guy and is super enthusiastic about this launch. But... my happy plates: Let us send you a 'curated' meal plan. Includes paleo and smoothie plans. Plans as low as $5/month! It's a list of recipes with the combined shopping list for the week. That's it. It promises to allow you to spend "less time searching Pinterest for recipes, more time snuggled up with the family."
|
|
#
?
Apr 9, 2016 20:21
|
|
|
If he varies them and keeps them simple, has some seasonal stuff, I could see people paying $5/mo for it. Then he does an Instacart tie in for one-click delivery of everything you need for the week!
|
|
#
?
Apr 9, 2016 20:28
|
|
|
Meal plans/shopping list subscriptions aren't a new thing. They're not even that bad of an idea. I think I might do one if I was a stay at home parent and thus had more time to cook.
|
|
#
?
Apr 9, 2016 20:30
|
|
|
Lucy Heartfilia posted:Truck convoys. 1 driver in the front truck and ten trucks without drivers that follow it. This Wednesday a Dutch government challenge to demonstrate a similar tech was succesfully completed. They had trucks from several brands follow each other very closely and brake/accelerate synchronized. If a driver is controlling eleven trucks they deserve eleven times their regular pay.
|
|
#
?
Apr 9, 2016 20:44
|
|
|
Unguided posted:If a driver is controlling eleven trucks they deserve eleven times their regular pay. This, but unironically.
|
|
#
?
Apr 9, 2016 20:57
|
|
|
Cicero posted:Meal plans/shopping list subscriptions aren't a new thing. They're not even that bad of an idea. I think I might do one if I was a stay at home parent and thus had more time to cook. Yeah it's not my thing, but it basically sounds like Plated or Blue Apron without the massive supply chain
|
|
#
?
Apr 9, 2016 21:05
|
|
|
Prof. Moriarty posted:A friend of a friend just launched his start-up. I feel a little guilty for posting it here because he seems like a nice enough guy and is super enthusiastic about this launch. But... There are a whole lot of these. I even just made up an outline for a curated meal app about an hour ago. It's actually a pretty good idea because the number one issue people have is not eating right, and the number one reason why they don't eat right is that they don't know what's right to eat. It's just such a simple idea that there's a bunch of competition.
|
|
#
?
Apr 9, 2016 22:19
|
|
|
It also seems like churn rate would be awful though
|
|
#
?
Apr 9, 2016 23:11
|
|
|
lancemantis posted:It also seems like churn rate would be awful though food for thought
|
|
#
?
Apr 9, 2016 23:14
|
|
|
Prof. Moriarty posted:A friend of a friend just launched his start-up. I feel a little guilty for posting it here because he seems like a nice enough guy and is super enthusiastic about this launch. But... So he's doing a CookSmarts clone that's a few bucks cheaper per month. Not a terrible idea, as it has obviously worked out for them.
|
|
#
?
Apr 9, 2016 23:15
|
|
|
Seems risky, some competitor could start offering "free" meal plans, trying to edge out the competition by focusing their entire business plan around Amazon referrals. Worse yet, suppose Amazon themselves caught on to the idea and started suggesting recipes to push premade grocery lists.
|
|
#
?
Apr 9, 2016 23:52
|
|
|
on the left posted:You are 100% wrong on this. I have talked with product planners from the Big 3 US Automakers and they all plan on having a high-end self-driving car for 2020, and expect the technology to be throughout their product lines by 2025. poo poo, I'd take that bet. If they solved the case handling in the next ten minutes it'd take at least 5 years to clear the legal hurdles that will come with the question of who is actually responsible for the car's behavior once the first one kills someone.
|
|
#
?
Apr 10, 2016 04:04
|
|
|
Not really, the manufacturers will probably set up a trade group and buy or partner with a large insurance company, passing the costs onto the end user. It might be a bit more complicated if the car had both automatic and manual modes, as there would probably need to be seperate insurance for each mode, and a definite way of proving what mode the car was in at any given time.
|
|
#
?
Apr 10, 2016 04:20
|
|
|
|
| # ? May 22, 2024 18:18 |
|
|
Liquid Communism posted:If they solved the case handling in the next ten minutes it'd take at least 5 years to clear the legal hurdles that will come with the question of who is actually responsible for the car's behavior once the first one kills someone. This is actually a somewhat solved problem. Volvo, for example, has already just said that they'll take responsibility directly for accidents involving any sort of autonomous mode in their cars. If self-driving cars ever become commonplace, it's going to be because manufacturers are confident enough in their systems to just accept full liability and bypass the need for complex legislation. Paradoxish fucked around with this message at 04:42 on Apr 10, 2016 |
|
#
?
Apr 10, 2016 04:39
|
|
